Blood vessel contour recognition method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses a blood vessel contour recognition method, device and equipment and a storage medium, and the method comprises the steps: obtaining a blood vessel ultrasonic image sequence, and carrying out the cross section blood vessel segmentation of each blood vessel ultrasonic image, so as to obtain a plurality of original blood vessel contour maps; identifying contour points in each original blood vessel contour map; respectively determining the uncertainty of each contour point in each original blood vessel contour map; and retaining or adjusting the position corresponding to the contour point according to the uncertainty so as to obtain the target blood vessel contour. According to the technical scheme, the problem that in the prior art, when the blood vessel contour in the IVUS image is recognized, the recognition accuracy is low is solved, the positions of the corresponding contour points can be adjusted based on the uncertainty of the contour points in the image, and the blood vessel contour recognition accuracy is improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of image processing technology, and in particular to a blood vessel contour recognition method, device, equipment and storage medium. Background Art

[0002] Imaging technology plays an important role in evaluating vascular anatomy, lesion characteristics and guiding surgical procedures. Among them, intravascular ultrasound (IVUS) introduces an ultrasound probe into a guidewire and places an ultrasound imaging device inside the coronary artery. It can provide cross-sectional and longitudinal images, allowing doctors to make detailed assessments of lesions inside the coronary artery. However, due to the resolution limitations of IVUS images themselves, the vascular contours may not be clear or difficult to distinguish in the image.

[0003] At present, the commonly used strategies and methods for the problem of vascular contours in IVUS images are: traditional image processing technology: through threshold segmentation, grayscale gradient-based and edge detection-based methods, manually design algorithm processes and related parameters to complete the extraction of vascular boundaries; data-driven segmentation methods: such as deep learning-based technologies, experienced clinicians annotate existing images to help machine learning models identify vascular contours.

[0004] A major difficulty faced by existing methods is that when the IVUS image itself appears blurred, it is difficult to maintain a good recognition accuracy when identifying the vascular contour. Summary of the invention

[0005] The embodiments of the present invention provide a blood vessel contour recognition method, apparatus, device and storage medium, which can adjust the positions of corresponding contour points based on the uncertainty of contour points in an image and improve the accuracy of blood vessel contour recognition.

[0006] In a first aspect, an embodiment of the present invention provides a method for recognizing a blood vessel contour, the method comprising:

[0007] Acquire a sequence of vascular ultrasound images, and perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images;

[0008] Identifying contour points in each of the original blood vessel contour images;

[0009] Determine the uncertainty of each contour point in each of the original blood vessel contour images respectively;

[0010] The positions of the corresponding contour points are retained or adjusted according to the uncertainty, thereby obtaining the target blood vessel contour.

[0011] In a second aspect, an embodiment of the present invention provides a blood vessel contour recognition device, the device comprising:

[0012] A cross-sectional segmentation module is used to obtain a sequence of vascular ultrasound images, and to perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images;

[0013] A contour point recognition module, used for recognizing contour points in each of the original blood vessel contour images;

[0014] An uncertainty determination module, used to respectively determine the uncertainty of each contour point in each of the original blood vessel contour images;

[0015] The contour adjustment module is used to retain or adjust the position of the corresponding contour point according to the uncertainty, so as to obtain the target blood vessel contour.

[0016] In a third aspect, an embodiment of the present invention provides a computer device, the computer device comprising:

[0017] one or more processors;

[0018] A memory for storing one or more programs;

[0019] When the one or more programs are executed by the one or more processors, the one or more processors implement the blood vessel contour recognition method described in any embodiment.

[0020] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the blood vessel contour recognition method described in any embodiment.

[0021] The technical solution provided by the embodiment of the present invention obtains a sequence of vascular ultrasound images, performs cross-sectional vascular segmentation on each vascular ultrasound image to obtain multiple original vascular contour maps; identifies contour points in each of the original vascular contour maps; determines the uncertainty of each contour point in each of the original vascular contour maps; retains or adjusts the position of the corresponding contour point based on the uncertainty, thereby obtaining the target vascular contour. The technical solution of the embodiment of the present invention solves the problem of low recognition accuracy when the prior art recognizes vascular contours in IVUS images, and can adjust the position of the corresponding contour point based on the uncertainty of the contour point in the image to improve the accuracy of vascular contour recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of a blood vessel contour recognition method provided by an embodiment of the present invention;

[0023] Figure 2is a flow chart of another blood vessel contour recognition method provided by an embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of original polar coordinate data provided by an embodiment of the present invention;

[0025] Figure 4 is a schematic diagram of another type of original polar coordinate data provided by an embodiment of the present invention;

[0026] Figure 5 is a structural schematic diagram of a blood vessel contour recognition device provided by an embodiment of the present invention;

[0027] Figure 6 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Figure 1 This is a flow chart of a blood vessel contour recognition method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of recognizing blood vessel contours in blood vessel ultrasound images. The method can be executed by a blood vessel contour recognition device, which can be implemented by software and / or hardware.

[0030] like Figure 1 As shown, the blood vessel contour recognition method includes the following steps:

[0031] S110 , acquiring a sequence of vascular ultrasound images, and performing cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images.

[0032] Among them, the vascular ultrasound image sequence may be a sequence including multiple vascular ultrasound images. Specifically, the vascular ultrasound image sequence may be obtained based on multiple images uploaded by an ultrasound probe. The original vascular contour map may be a vascular ultrasound image obtained after image preprocessing. Specifically, each vascular ultrasound image in the vascular ultrasound image sequence may be segmented into cross-section vessels based on an image frame of a preset size, thereby obtaining multiple original vascular contour maps. The technical solution of the implementation of the present invention requires the recognition of the vascular contour in the original vascular contour map.

[0033] S120: Identify contour points in each of the original blood vessel contour images.

[0034] Wherein, each original blood vessel contour image is input into a pre-trained target blood vessel contour recognition model to obtain contour points in the original blood vessel contour image.

[0035] Furthermore, when the contour points in the image are identified only by the target blood vessel contour recognition model, the contour point recognition accuracy may be low. In particular, when the original blood vessel contour image is fuzzy, the accuracy of the contour points identified based on the target blood vessel contour recognition model will be more difficult to guarantee. Therefore, the recognition accuracy of each contour point can be analyzed later, and then the position of the corresponding contour point can be adjusted to improve the recognition accuracy of the blood vessel contour.

[0036] S130, respectively determining the uncertainty of each contour point in each of the original blood vessel contour images.

[0037] The uncertainty can be the degree of uncertainty that the identified contour point is a real blood vessel contour point. Specifically, the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a blood vessel contour point can be calculated respectively, and then the uncertainty can be obtained based on the calculated probability value.

[0038] S140 , retaining or adjusting the position of the corresponding contour point according to the uncertainty, thereby obtaining the target blood vessel contour.

[0039] The target blood vessel contour may be the blood vessel contour in the original blood vessel contour map. The target blood vessel contour may be obtained from all contour points in the original blood vessel contour map. Therefore, it is very important to ensure the recognition accuracy of the contour points.

[0040] Specifically, a corresponding preset threshold may be set. When the uncertainty of a contour point is less than the preset threshold, the position of the current contour point is retained; when the uncertainty of a contour point is greater than or equal to the preset threshold, the position of the corresponding contour point is adjusted.

[0041] When the uncertainty of a contour point is less than a preset threshold, it indicates that the contour point is a true blood vessel contour point with high credibility, so the position of the contour point may not be adjusted and the current position of the contour point may be retained.

[0042] When the uncertainty of the contour point is equal to or greater than the preset threshold, the credibility of the contour point as a true blood vessel contour point is low, and thus the position of the contour point may be adjusted later. Specifically, when the uncertainty of the current contour point is equal to or greater than the preset threshold, a reference blood vessel contour point in an original blood vessel contour map similar to the current original blood vessel contour map may be determined first, and then the position may be adjusted based on the positional association between the current contour point and each reference blood vessel contour point.

[0043] The technical solution of the embodiment of the present invention can improve the credibility of each processed contour point and improve the accuracy of blood vessel contour recognition by retaining or adjusting the position of the corresponding contour point according to the uncertainty.

[0044] Optionally, the corresponding target blood vessel contour may be determined based on the contour points in each original blood vessel contour map. The target blood vessel contour may be a three-dimensional blood vessel contour graphic corresponding to the blood vessel ultrasound image sequence. Specifically, after determining the contour points in each original blood vessel contour map, the positions of the contour points may be fitted to obtain the target blood vessel contour. By determining the corresponding target blood vessel contour through the contour points in the original blood vessel contour map, a more three-dimensional blood vessel contour graphic may be obtained, thereby increasing the visibility of the identified blood vessel contour.

[0045] The technical solution provided by the embodiment of the present invention obtains a sequence of vascular ultrasound images, performs cross-sectional vascular segmentation on each vascular ultrasound image to obtain multiple original vascular contour maps; identifies contour points in each original vascular contour map; determines the uncertainty of each contour point in each original vascular contour map; retains or adjusts the position of the corresponding contour point based on the uncertainty, thereby obtaining the target vascular contour. The technical solution of the embodiment of the present invention solves the problem of low recognition accuracy when the prior art recognizes the vascular contour in IVUS images, and can adjust the position of the corresponding contour point based on the uncertainty of the contour point in the image to improve the accuracy of vascular contour recognition.

[0046] Figure 2 This is another flow chart of a blood vessel contour recognition method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scene of recognizing blood vessel contours in blood vessel ultrasound images. Based on the above embodiment, this embodiment further explains how to identify contour points in each original blood vessel contour map; how to determine the uncertainty of each contour point in each original blood vessel contour map; and how to retain or adjust the position of the corresponding contour point based on the uncertainty. The device can be implemented by software and / or hardware, and integrated into a computer device with application development function.

[0047] like Figure 2 As shown, the blood vessel contour recognition method includes the following steps:

[0048] S210 , acquiring a sequence of vascular ultrasound images, and performing cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images.

[0049] The vascular ultrasound image sequence may be a sequence including multiple vascular ultrasound images. Specifically, the vascular ultrasound image sequence may be obtained based on multiple images uploaded by an ultrasound probe. The original vascular contour map may be a vascular ultrasound image obtained after image preprocessing. Specifically, each vascular ultrasound image in the vascular ultrasound image sequence may be segmented into a cross-section vessel based on an image frame of a preset size, thereby obtaining multiple original vascular contour maps.

[0050] For example, a large amount of clinical real IVUS image data can be obtained through IVUS imaging equipment and preprocessed. Usually, the size of IVUS image data is 512*512*N, where N is defined as the number of layers of IVUS images. For each layer, we define it as a cross section with an image size of 512*512. Through data annotation and other schemes, a convolutional neural network can be used to segment blood vessels for each cross section to obtain an image of the same size of 512*512. The value P of each point in the image is i ∈(0,1), represents the probability value that the point is a blood vessel. The technical solution of the embodiment of the present invention needs to identify the blood vessel contour in the original blood vessel contour map later.

[0051] S220 , respectively inputting each of the original blood vessel contour images into a pre-trained target blood vessel contour recognition model to obtain contour points in the original blood vessel contour image.

[0052] The target blood vessel contour recognition model may be a model for recognizing blood vessel contours in an image. Specifically, a convolutional neural network may be used as an initial blood vessel contour recognition model, and then the initial blood vessel contour recognition model may be trained based on corresponding samples to obtain a target blood vessel contour recognition model. After the original blood vessel contour image is input into the target blood vessel contour recognition model, the model may recognize contour points of the blood vessel contour in the original blood vessel contour image.

[0053] Furthermore, when the contour points in the image are identified only by the target blood vessel contour recognition model, the contour point recognition accuracy may be low. In particular, when the original blood vessel contour image is fuzzy, the accuracy of the contour points identified based on the target blood vessel contour recognition model will be more difficult to guarantee. Therefore, the recognition accuracy of each contour point can be analyzed later, and then the position of the corresponding contour point can be adjusted to improve the recognition accuracy of the blood vessel contour.

[0054] S230 , mapping each contour point in the original blood vessel contour image into a preset polar coordinate system to obtain original polar coordinate data corresponding to each original blood vessel contour image.

[0055] The preset polar coordinate system may be a preset polar coordinate system for analyzing the accuracy of contour point recognition. Specifically, the preset polar coordinate system takes the center of the original blood vessel contour image as the origin, the angle as the horizontal coordinate, and the distance from the contour point to the boundary of the blood vessel cross-sectional image as the vertical coordinate. The original polar coordinate data may be the mapping data of the original blood vessel contour image in the preset polar coordinate system. Specifically, each contour point in the original blood vessel contour image may be mapped in the preset polar coordinate system to obtain the original polar coordinate data corresponding to each original blood vessel contour image.

[0056] S240 , determining uncertainty according to the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a blood vessel contour point.

[0057] Among them, the uncertainty can be the degree of uncertainty that the identified contour point is the real blood vessel contour point. Considering the morphological characteristics of blood vessels as circular, in order to more conveniently obtain the contour, we convert the segmentation results into polar coordinates for uncertainty estimation. Specifically, the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a blood vessel contour point can be calculated respectively, and then the uncertainty can be obtained based on the calculated probability value. Exemplarily, informatics metrics such as entropy can be used to measure the uncertainty of the boundary, and the calculation formula for uncertainty is as follows:

[0058] H(I)=-∑p(i)logp(i)

[0059] Where H(I) represents the uncertainty, and p(i) represents the probability that contour point i is a blood vessel contour point.

[0060] The embodiment of the present invention maps each contour point in the original blood vessel contour map into a preset polar coordinate system, and can find a reference coordinate point that has a longitudinal morphological relationship with the current coordinate point from an angle, so as to facilitate the subsequent adjustment of the position of the current coordinate point based on the reference coordinate point and adjust the efficiency of the coordinate point position adjustment.

[0061] For example, Figure 3 Schematic diagram of original polar coordinate data provided by an embodiment of the present invention. Figure 3 As shown, when the uncertainty of the contour points is small, the blood vessel boundary is clear and the distribution of the mapping results at the edge is sharp.

[0062] For example, Figure 4 Schematic diagram of another type of original polar coordinate data provided by an embodiment of the present invention. Figure 4 As shown in the figure, when the uncertainty of the contour points is large, the boundary is not clear and the distribution at the edge is diffuse.

[0063] Subsequently, the determined uncertainty may be compared with a preset threshold. When the uncertainty of the contour point is less than the preset threshold, step S250 is executed; when the uncertainty of the contour point is greater than or equal to the preset threshold, step S260 is executed.

[0064] S250: When the uncertainty of the contour point is less than a preset threshold, retain the position of the current contour point.

[0065] The preset threshold may be a preset reference threshold for evaluating uncertainty. When the uncertainty of a contour point is less than the preset threshold, it indicates that the contour point is a true blood vessel contour point with high credibility, so the position of the contour point may not be adjusted and the current position of the contour point may be retained.

[0066] S260. When the uncertainty of the contour point is greater than or equal to a preset threshold, obtain reference positions corresponding to the position of the current contour point on at least two non-current original blood vessel contour images, and adjust the position of the current contour point according to the relationship between the position of the current contour point and the longitudinal morphology of the blood vessel presented by all reference positions.

[0067] When the uncertainty of a contour point is equal to or greater than a preset threshold, the confidence that the contour point is a true vascular contour point is low, and thus the position of the contour point may be adjusted later. Specifically, reference positions corresponding to the position of the current contour point on at least two non-current original vascular contour images may be obtained, and the position of the current contour point may be adjusted based on the relationship between the position of the current contour point and the longitudinal morphology of the vascular presented by all reference positions.

[0068] Among them, the current contour point position may be the position of the current contour point whose uncertainty is less than a preset threshold. The reference position may be a contour point position associated with the current contour point position in a non-current original blood vessel contour map. Furthermore, the position of the current contour point may be adjusted accordingly based on the positional association relationship of the blood vessel contour point presented at the reference position. The technical solution of the embodiment of the present invention can improve the credibility of each adjusted contour point and improve the accuracy of blood vessel contour recognition by adjusting the position of the corresponding contour point based on the uncertainty. At the same time, it can also reduce the workload of doctors and improve the overall automation of auxiliary diagnosis software.

[0069] Specifically, the coordinate points to be updated in the current original vascular contour map and the corresponding original polar coordinate data can be first determined based on the uncertainty of the current contour point; at least two original vascular contour maps that are similar to the current original vascular contour map are obtained as reference vascular contour maps, and mapping data of the reference vascular contour map in a preset polar coordinate system are obtained to obtain reference polar coordinate data; based on the coordinate information of the coordinate points to be updated, the reference coordinate points of the reference polar coordinate data are determined; and based on the longitudinal morphological relationship between the coordinate points to be updated and the reference coordinate points, the position of the current contour point is adjusted.

[0070] Among them, the current original blood vessel contour map may be the original blood vessel contour map corresponding to the current contour point. The coordinate point to be updated may be the coordinate point corresponding to the current contour point in the original polar coordinate data. The reference blood vessel contour map may be an image used to adjust the position of the current contour point. Specifically, at least two original blood vessel contour maps similar to the current original blood vessel contour map may be used as reference blood vessel contour maps. The reference polar coordinate data may be mapping data of the reference blood vessel contour map in a preset polar coordinate system. The reference coordinate point may be a coordinate point in the reference polar coordinate data used for subsequent adjustment of the position of the coordinate point to be updated. In order to increase the credibility of the reference coordinate point, the uncertainty of the contour point corresponding to the reference coordinate point may be limited to be less than a preset threshold. Specifically, according to the coordinate information of the coordinate point to be updated, a coordinate point in the reference coordinate data having a position association relationship with the coordinate point to be updated may be determined as a reference coordinate point, and then the position of the current contour point may be adjusted according to the longitudinal morphological relationship between the coordinate point to be updated and the reference coordinate point.

[0071] Specifically, when determining the reference coordinate point, the coordinate point with the same horizontal coordinate as the coordinate point to be updated in the reference polar coordinate data can be used as the reference coordinate point. The horizontal coordinate in the preset polar coordinate system is an angle, which can be understood as each coordinate point corresponding to a contour point at a certain angle in the original image. Therefore, the blood vessel contour coordinate point with the same horizontal coordinate as the coordinate point to be updated is used as the reference coordinate point, and the positional relationship between the blood vessel contour coordinate points at the same angle as the coordinate point to be updated can be analyzed, and then the position of the coordinate point to be updated can be adjusted to improve the accuracy of determining the blood vessel contour point.

[0072] Furthermore, when adjusting the position of the current contour point based on the longitudinal morphological relationship between the coordinate point to be updated and the reference coordinate point, the longitudinal coordinate of the coordinate point to be updated can be adjusted according to the coordinate position relationship of each reference coordinate point to determine the position of the target coordinate point; according to the position of the target coordinate point, the position of the contour point corresponding to the coordinate point to be updated is adjusted.

[0073] Among them, the position coordinate relationship can be a position association relationship between each reference coordinate point. Specifically, the position relationship between each reference coordinate point can be analyzed to determine the coordinate position relationship. The target coordinate point position can be the coordinate point position after the position of the coordinate point to be updated is adjusted. Specifically, the ordinate of the coordinate point to be updated can be adjusted according to the coordinate position relationship between the reference coordinate points to obtain the target coordinate point position. After determining the position of the target coordinate point, the position of the contour point corresponding to the coordinate point to be updated can be adjusted accordingly according to the mapping relationship between the image and the preset polar coordinate system. By taking the blood vessel contour coordinate point with the same horizontal coordinate as the coordinate point to be updated as the reference coordinate point, and then determining the coordinate position relationship according to the reference coordinate point, the vertical coordinate of the coordinate point to be updated can be adjusted, and the adjustment amount of the position point of the coordinate point can be reduced to an indicator of the vertical coordinate, which can reduce the amount of calculation for determining the position of the target coordinate point and improve the efficiency and accuracy of the position adjustment.

[0074] Optionally, position fitting may be performed on each reference coordinate point to obtain a contour coordinate fitting curve; and the ordinate of the contour coordinate point to be updated may be adjusted according to the contour coordinate fitting curve to obtain the position of the target coordinate point.

[0075] Among them, the contour coordinate fitting curve can be a fitting curve representing the positional relationship between the reference coordinate points. Specifically, the position points of each reference coordinate point can be fitted to obtain the contour coordinate fitting curve. Then, the position of the target coordinate point can be determined according to the contour coordinate fitting curve. For example, when the expression of the contour coordinate fitting curve is Y=Ax+b, the horizontal coordinate of the coordinate point to be updated can be substituted into the expression to obtain the position of the target coordinate point. Then, the coordinate point to be updated can be adjusted to the position of the target coordinate point to achieve the position adjustment of the coordinate point to be updated. By fitting the position of each reference coordinate point to obtain the contour coordinate fitting curve, the position change trend of the blood vessel contour points in each image can be obtained according to the contour coordinate fitting curve, and then the target adjustment position of the current coordinate point can be determined based on the position change trend, thereby improving the accuracy of the coordinate point position adjustment.

[0076] The technical solution provided by the embodiment of the present invention obtains a sequence of vascular ultrasound images, performs cross-sectional vascular segmentation on each vascular ultrasound image to obtain multiple original vascular contour images; inputs each original vascular contour image into a pre-trained target vascular contour recognition model to obtain contour points in the original vascular contour image; maps each contour point in the original vascular contour image in a preset polar coordinate system to obtain original polar coordinate data corresponding to each original vascular contour image; determines uncertainty based on the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a vascular contour point; when the uncertainty of the contour point is less than a preset threshold, retains the position of the current contour point; when the uncertainty of the contour point is greater than or equal to the preset threshold, obtains the reference position corresponding to the current contour point position on at least two non-current original vascular contour images, and adjusts the position of the current contour point according to the relationship between the position of the current contour point and the longitudinal morphology of the blood vessels presented by all reference positions. The technical solution of the embodiment of the present invention solves the problem of low recognition accuracy in the prior art when recognizing vascular contours in IVUS images, and can adjust the position of the corresponding contour point based on the uncertainty of the contour point in the image to improve the recognition accuracy of the vascular contour.

[0077] Figure 5 It is a structural schematic diagram of a blood vessel contour recognition device provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of recognizing blood vessel contours in blood vessel ultrasound images. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0078] like Figure 5 As shown, the blood vessel contour recognition device includes: a cross-section segmentation module 310 , a contour point recognition module 320 , an uncertainty determination module 330 and a contour adjustment module 340 .

[0079] Among them, the cross-sectional segmentation module 310 is used to obtain a sequence of vascular ultrasound images, and perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain multiple original vascular contour images; the contour point recognition module 320 is used to identify the contour points in each of the original vascular contour images; the uncertainty determination module 330 is used to determine the uncertainty of each contour point in each of the original vascular contour images; the contour adjustment module 340 is used to retain or adjust the position of the corresponding contour point according to the uncertainty, so as to obtain the target vascular contour.

[0080] The technical solution provided by the embodiment of the present invention obtains a sequence of vascular ultrasound images, performs cross-sectional vascular segmentation on each vascular ultrasound image to obtain multiple original vascular contour maps; identifies contour points in each of the original vascular contour maps; determines the uncertainty of each contour point in each of the original vascular contour maps; retains or adjusts the position of the corresponding contour point based on the uncertainty, thereby obtaining the target vascular contour. The technical solution of the embodiment of the present invention solves the problem of low recognition accuracy when the prior art recognizes vascular contours in IVUS images, and can adjust the position of the corresponding contour point based on the uncertainty of the contour point in the image to improve the accuracy of vascular contour recognition.

[0081] In an optional embodiment, the contour adjustment module 340 includes: a first adjustment submodule and a second adjustment submodule; wherein the first adjustment submodule is used to: when the uncertainty of the contour point is less than a preset threshold, retain the position of the current contour point; the second adjustment submodule is used to: when the uncertainty of the contour point is greater than or equal to the preset threshold, obtain the reference position corresponding to the current contour point position on at least two non-current original blood vessel contour images, and adjust the position of the current contour point according to the relationship between the position of the current contour point and the longitudinal morphology of the blood vessel presented by all reference positions.

[0082] In an optional embodiment, the uncertainty determination module 330 is specifically used to: map each contour point in the original blood vessel contour image in a preset polar coordinate system to obtain original polar coordinate data corresponding to each original blood vessel contour image; determine the uncertainty according to the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a blood vessel contour point; wherein the preset polar coordinate system takes the center of the original blood vessel contour image as the origin, the angle as the horizontal coordinate, and the distance from the contour point to the boundary of the blood vessel cross-sectional image as the vertical coordinate.

[0083] In an optional embodiment, the second adjustment submodule is specifically used to: determine the coordinate point to be updated in the current original vascular contour map and the corresponding original polar coordinate data according to the uncertainty of the current contour point; obtain at least two original vascular contour maps similar to the current original vascular contour map as reference vascular contour maps, and obtain the mapping data of the reference vascular contour map in the preset polar coordinate system to obtain reference polar coordinate data; determine the reference coordinate point of the reference polar coordinate data according to the coordinate information of the coordinate point to be updated; and adjust the position of the current contour point according to the longitudinal morphological relationship between the coordinate point to be updated and the reference coordinate point.

[0084] In an optional embodiment, the second adjustment submodule includes a coordinate point position determination unit and a position adjustment unit, wherein the coordinate point position determination unit is used to: adjust the vertical coordinate of the coordinate point to be updated according to the coordinate position relationship of each reference coordinate point to determine the target coordinate point position; the position adjustment unit is used to: adjust the position of the contour point corresponding to the coordinate point to be updated according to the target coordinate point position.

[0085] In an optional implementation, the coordinate point position determination unit is specifically used to: perform position fitting on each of the reference coordinate points to obtain a contour coordinate fitting curve; and adjust the vertical coordinate of the contour coordinate point to be updated according to the contour coordinate fitting curve to obtain the target coordinate point position.

[0086] In an optional implementation, the uncertainty determination module 330 is specifically configured to: input each of the original blood vessel contour images into a pre-trained target blood vessel contour recognition model to obtain contour points in the original blood vessel contour image.

[0087] In an optional implementation, the blood vessel contour recognition device further includes: a blood vessel contour determination module, configured to determine a corresponding target blood vessel contour according to contour points in each of the original blood vessel contour images.

[0088] The blood vessel contour recognition device provided in the embodiment of the present invention can execute the blood vessel contour recognition method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0089] Figure 6 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 6 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 6 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capability and can be configured in the blood vessel contour recognition device.

[0090] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0091] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0092] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0093] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0094] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0095] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 20. Figure 6 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0096] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the blood vessel contour recognition method provided by the embodiment of the present invention, the method comprising:

[0097] Acquire a sequence of vascular ultrasound images, and perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images;

[0098] Identifying contour points in each of the original blood vessel contour images;

[0099] Determine the uncertainty of each contour point in each of the original blood vessel contour images respectively;

[0100] The positions of the corresponding contour points are retained or adjusted according to the uncertainty, thereby obtaining the target blood vessel contour.

[0101] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the blood vessel contour recognition method provided by any embodiment of the present invention is implemented, including:

[0102] Acquire a sequence of vascular ultrasound images, and perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images;

[0103] Identifying contour points in each of the original blood vessel contour images;

[0104] Determine the uncertainty of each contour point in each of the original blood vessel contour images respectively;

[0105] The positions of the corresponding contour points are retained or adjusted according to the uncertainty, thereby obtaining the target blood vessel contour.

[0106] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0107] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0108] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0109] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0110] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0111] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A blood vessel contour recognition method, characterized in that: include: Acquire a sequence of vascular ultrasound images, and perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images; Identifying contour points in each of the original blood vessel contour images; Determine the uncertainty of each contour point in each of the original blood vessel contour images respectively; The positions of the corresponding contour points are retained or adjusted according to the uncertainty, thereby obtaining the target blood vessel contour.

2. The method according to claim 1, characterized in that The adjusting the position of the corresponding contour point according to the uncertainty includes: When the uncertainty of the contour point is less than a preset threshold, retaining the position of the current contour point; or, When the uncertainty of the contour point is greater than or equal to a preset threshold, obtain reference positions corresponding to the position of the current contour point on at least two non-current original blood vessel contour images, and adjust the position of the current contour point based on the relationship between the position of the current contour point and the longitudinal morphology of the blood vessel presented by all reference positions.

3. The method according to claim 2, characterized in that The step of respectively determining the uncertainty of each contour point in each of the blood vessel cross-sectional images comprises: Mapping each contour point in the original blood vessel contour image in a preset polar coordinate system to obtain original polar coordinate data corresponding to each original blood vessel contour image; Determining the uncertainty according to the probability that the coordinate point corresponding to each contour point in the original polar coordinate data is a blood vessel contour point; The preset polar coordinate system takes the center of the original blood vessel contour image as the origin, the angle as the abscissa, and the distance from the contour point to the boundary of the blood vessel cross-sectional image as the ordinate.

4. The method according to claim 3, characterized in that When the uncertainty of the contour point is greater than or equal to a preset threshold, obtaining reference positions corresponding to the position of the current contour point on at least two non-current original blood vessel contour images, and adjusting the position of the current contour point according to the relationship between the position of the current contour point and the longitudinal morphology of the blood vessel presented by all reference positions, including: Determine the coordinate points to be updated in the current original blood vessel contour image and the corresponding original polar coordinate data according to the uncertainty of the current contour point; Acquire at least two original blood vessel contour images that are similar to the current original blood vessel contour image as reference blood vessel contour images, and acquire mapping data of the reference blood vessel contour images in the preset polar coordinate system to obtain reference polar coordinate data; Determining a reference coordinate point of the reference polar coordinate data according to the coordinate information of the coordinate point to be updated; The position of the current contour point is adjusted according to the longitudinal morphological relationship between the coordinate point to be updated and the reference coordinate point.

5. The method according to claim 4, characterized in that The adjusting the position of the current contour point according to the longitudinal morphological relationship between the coordinate point to be updated and the reference coordinate point comprises: According to the coordinate position relationship of each of the reference coordinate points, the ordinate of the coordinate point to be updated is adjusted to determine the position of the target coordinate point; According to the position of the target coordinate point, the position of the contour point corresponding to the coordinate point to be updated is adjusted.

6. The method according to claim 5, characterized in that The adjusting the ordinate of the coordinate point to be updated according to the coordinate position relationship of each reference coordinate point to determine the position of the target coordinate point includes: Performing position fitting on each of the reference coordinate points to obtain a contour coordinate fitting curve; The ordinate of the to-be-updated contour coordinate point is adjusted according to the contour coordinate fitting curve to obtain the target coordinate point position.

7. The method according to claim 1, characterized in that The identifying of the contour points in each of the original blood vessel contour images comprises: Each of the original blood vessel contour images is input into a pre-trained target blood vessel contour recognition model to obtain contour points in the original blood vessel contour image.

8. The method according to claim 1, characterized in that The method further comprises: According to the contour points in each of the original blood vessel contour images, the corresponding target blood vessel contour is determined.

9. A blood vessel contour recognition device, characterized in that: The device comprises: A cross-sectional segmentation module is used to obtain a sequence of vascular ultrasound images, and to perform cross-sectional vascular segmentation on each vascular ultrasound image to obtain a plurality of original vascular contour images; A contour point recognition module, used for recognizing contour points in each of the original blood vessel contour images; An uncertainty determination module, used to respectively determine the uncertainty of each contour point in each of the original blood vessel contour images; The contour adjustment module is used to retain or adjust the position of the corresponding contour point according to the uncertainty, so as to obtain the target blood vessel contour.

10. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the blood vessel contour recognition method as described in any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the blood vessel contour recognition method as described in any one of claims 1 to 8 is implemented.

Images